Jump to content

Mimetic interpolation

From Wikipedia, the free encyclopedia

In mathematics, mimetic interpolation is a method for interpolating differential forms. In contrast to other interpolation methods, which estimate a field at a location given its values on neighboring points, mimetic interpolation estimates the field's -form given the field's projection on neighboring grid elements. The grid elements can be grid points as well as cell edges or faces, depending on .

Mimetic interpolation is particularly relevant in the context of vector and pseudo-vector fields as the method conserves line integrals and fluxes, respectively.

Interpolation of integrated forms

[edit]

Let be a differential -form, then mimetic interpolation is the linear combination

where is the interpolation of , and the coefficients represent the strengths of the field on grid element . Depending on , can be a node (), a cell edge (), a cell face () or a cell volume (). In the above, the are the interpolating -forms, which are centered on and decay away from in a way similar to the tent functions. Examples of are the Whitney forms[1][2] for simplicial meshes in dimensions.

An important advantage of mimetic interpolation over other interpolation methods is that the field strengths are scalars and thus coordinate system invariant.

Interpolating forms

[edit]

In many cases it is desirable for the interpolating forms to pick the field's strength on particular grid elements without interference from other . This allows one to assign field values to specific grid elements, which can then be interpolated in-between. A common case is linear interpolation for which the interpolating functions (-forms) are zero on all nodes except on one, where the interpolating function is one. A similar construct can be applied to mimetic interpolation

That is, the integral of is zero on all cell elements, except on where the integral returns one. For this amounts to where is a grid point. For the integral is over edges and hence the integral is zero expect on edge . For the integral is over faces and for over cell volumes.

Conservation properties

[edit]

Mimetic interpolation respects the properties of differential forms. In particular, Stokes' theorem

is satisfied with denoting the interpolation of . Here, is the exterior derivative, is any manifold of dimensionality and is the boundary of . This confers to mimetic interpolation conservation properties, which are not generally shared by other interpolation methods.

Commutativity between the interpolation operator and the exterior derivative

[edit]
The figure depicts the commutativity condition between interpolation and exterior derivative operators.
De Rham complex. Top: the spaces of differential forms in three dimensions. Bottom: the corresponding discretized versions of the differential forms obtained after interpolation. The commutativity condition ensures that the dashed and dash-dotted paths give the same result for each de Rham cell.

To be mimetic, the interpolation must satisfy

where is the interpolation operator of a -form, i.e. . In other words, the interpolation operators and the exterior derivatives commute.[3] Note that different interpolation methods are required for each type of form (), . The above equation is all that is needed to satisfy Stokes' theorem for the interpolated form

Other calculus properties derive from the commutativity between interpolation and . For instance, ,

The last step gives zero since when integrated over the boundary .

Projection

[edit]

The interpolated is often projected onto a target, -dimensional, oriented manifolds ,For the target is a point, for it is a line, for an area, etc.

Applications

[edit]

Many physical fields can be represented as -forms. When discretizing fields in numerical modeling, each type of -form often acquires its own staggering in accordance with numerical stability requirements, e.g. the need to prevent the checkerboard instability.[4] This led to the development of the exterior finite element[5] and discrete exterior calculus methods, both of which rely on a field discretization that are compatible with the field type.

The table below lists some examples of physical fields, their type, their corresponding form and interpolation method, as well as software that can be leveraged to interpolate, remap or regrid the field:

field example field type k-form equivalent target staggering Interpolation method (example) example of software
temperature scalar 0-form point nodal bilinear, trilinear ESMF[6]
electric field vector 1-form line edge edge MINT[7]
magnetic field pseudo-vector 2-form area face face MINT
density pseudo-scalar 3-form volume cell area weighted, conservative SCRIP,[8] ESMF

Example

[edit]
Nodes and edges are indexed in the counterclockwise direction. Edge 0 goes from node 0 to 1, edge 1 from node 1 to 2, edge 2 goes from node 3 to 2 and edge 3 goes from node 0 to 3
Indexing of nodes and edges for a quadrilateral cell, as used in the example. The edges are chosen to point in the east and north directions.

Consider quadrilateral cells in two dimensions with their node indexed in the counterclockwise direction. Further, let and be the parametric coordinates of each cell (). Then

are the bilinear interpolating forms of in the unit square (). The corresponding edge interpolating forms[9][10] are

Visual representation of the edge interpolating forms.
The four vector fields which are dual to the four edge interpolating forms attached to a quadrilateral cell. The vector fields are strongest on their supporting edge and decrease to zero towards the opposite edge. Note how the vector fields bend to enforce perpendicularity with respect to edges that are adjacent to the supporting edge.

were we assumed the edges to be indexed in counterclockwise direction and with the edges pointing to the east and north. At lowest order, there is only one interpolating form for ,

where is the wedge product.

We can verify that the above interpolating forms satisfy the mimetic conditions and . Take for instance,

where , , and are the field values evaluated at the corners of the quadrilateral in the unit square space. Likewise, we have

with , , being the 1-form projected onto edge . Note that is also known as the pullback. If is the map that parametrizes edge , , , then where the integration is performed in space. Consider for instance edge , then with and denoting the start and points. For a general 1-form , one gets .

References

[edit]
  1. ^ Whitney, Hassler (1957). Geometric Integration Theory. Dover Books on Mathematics.
  2. ^ Hiptmair, R (2022-06-12). "Higher Order Whitney Forms". Progress in Electromagnetics Research. 32: 271–299. doi:10.2528/PIER00080111.
  3. ^ Pletzer, Alexander; Behrens, Erik; Little, Bill (2022-06-27). "MINT". Proceedings of the Platform for Advanced Scientific Computing Conference. Basel Switzerland: ACM. pp. 1–7. doi:10.1145/3539781.3539786. ISBN 978-1-4503-9410-9.
  4. ^ Trottenberg, Ulrich; Oosterlee, Cornelius W.; Schüller, Anton (2001). Multigrid. Academic Press. p. 314.
  5. ^ Arnold, Douglas N.; Falk, Richard S.; Winther, Ragnar (2022-06-12). "Finite element exterior calculus, homological techniques, and applications". Acta Numerica. 15: 1–155. doi:10.1017/S0962492906210018. S2CID 122763537.
  6. ^ "Earth System Modeling Framework Regridding".
  7. ^ "Mimetic Interpolation on the Sphere". GitHub. 4 March 2022. Retrieved 2022-06-09.
  8. ^ "SCRIP". GitHub. 11 April 2022. Retrieved 2022-06-09.
  9. ^ Pletzer, Alexander; Fillmore, David (2015-12-01). "Conservative interpolation of edge and face data on n dimensional structured grids using differential forms". Journal of Computational Physics. 302: 21–40. Bibcode:2015JCoPh.302...21P. doi:10.1016/j.jcp.2015.08.029. ISSN 0021-9991.
  10. ^ Pletzer, Alexander; Hayek, Wolfgang (2019-01-01). "Mimetic Interpolation of Vector Fields on Arakawa C/D Grids". Monthly Weather Review. 147 (1): 3–16. Bibcode:2019MWRv..147....3P. doi:10.1175/MWR-D-18-0146.1. ISSN 0027-0644. S2CID 125214770.